Pediatric Exercise Science, 1995,7 , 94-102
O 1995 Human Kinetics Publishers, Inc.
Use of Rating of Perceived Exertion (RPE)
to Prescribe Exercise intensity
for Wheelchair-Bound Children and Adults
Dianne S. Ward, Oded Bar-Or,
Patti Longmuir, and Karen Smith
Seventeen individuals (ages 11-30 years), all wheelchair users, were classified as active or sedentary. Peak mechanical power, heart rate (HR), and
rating of perceived exertion (RPE) were determined during continuous, incremental all-out arm ergometry. Subjects were asked to wheel on an oval track
at prescribed speeds, and one month later they repeated this task. All subjects
could distinguish among prescriptions, as judged from HR and wheeling
velocities. However, the active subjects chose higher speeds (by 0.81.3 m/s), a wider range of speeds, and could better distinguish among sequential RPE levels than did the sedentary subjects. All subjects chose wheeling
velocities higher than expected from their originally established HR-onRPE regression. One-month retention was high and similar between groups.
Individuals who use wheelchairs can discriminate among wheeling intensities
as prescribed using the RPE scale and have excellent retention for at least
one month.
Self-estimation of exertion using some type of rating scale has been employed for many years as supplementary information to data obtained during
graded exercise testing (2, 8). Recently, subjective rating scales such as Borg's
6 2 0 category rating of perceived exertion (RPE) scale (7) have been used as a
method of prescribing intensity (see Birk and Birk [6] for review). Based on a
diagnostic fitness evaluation, a "window" of RPE numbers (e.g., 12-15) is
prescribed as the target intensity level, and the individual attempts to maintain
exercise within this range. The use of RPE to gauge exercise intensity has been
recognized by the American College of Sports Medicine (ACSM) as a useful
technique, particularly when used in conjunction with heart rate (1).
Research studies on the effectiveness of RPE as a tool to gauge exercise
intensity generally have used adults (9, 12, 13, 16) or able-bodied children (17,
18, 19). There is no information to date on whether a disabled population,
Dianne S. Ward is with the Dept. of Exercise Science at the University of South
Carolina, Columbia, SC 29208. Oded Bar-Or and Karen Smith are with the Children's
Exercise and Nutrition Centre, McMaster University, Sanatorium Rd., Hamilton, ON
Canada L8N 325. Patti Longmuir is with Variety Village, 3701 Danforth Ave., Scarborough, ON Canada M l N 2G2.
Rating of Perceived Exertion - 95
specifically the wheelchair-bound, can employ RPE effectively for exercise training. Thus, it was the purpose of this study to determine if children and adults
who use wheelchairs can accurately pace their wheeling around a track using
intensities prescribed to them as numbers on the Borg RPE category scale. An
additional purpose was to determine whether this ability differs between active
and inactive individuals, and whether it is retained one month later.
Methods
Subjects
Seventeen volunteers, ages 11-30 years, were recruited from organizations for
the disabled or sports clubs, qualified for participation, and were subsequently
divided into groups based on physical activity status. Activity status was determined by questionnaire (modified from Bar-Or and Ward [3], pp. 151-168). An
active subject was defined as one who participated in exercise workouts more
than three to four times per week and that lasted 15-20 min per session. Sedentary
subjects were those whose activity patterns occurred less than three times per
week and that lasted less than 20 min per session. Most subjects in the active
category were involved in some form of regular sport competition. All subjects
had disabilities requiring use of a wheelchair. Disabilities included spina bifida,
cerebral palsy, and traumatic paraplegia. Level of spinal lesion was of no consequence in subject selection as long as the individual could wheel his or her own
nonpower chair.
Design and Procedures
Session 1. This session was devoted to orientation and collection of
baseline information. Arm span was taken to reflect body height using a measuring
tape permanently affixed to a length of wood. Skinfold thickness (mean of
three trials) was measured with Harpenden calipers at four sites: biceps, triceps,
subscapular, and suprailiac (1 1). Body weight was measured with a Mott electronic scale accurate to 20 g (Model LC2424).
Following the anthropometric measurements, the RPE 6-20 scale was
introduced and then used within a progressive, continuous, maximal arm ergometer test as described by Bar-Or and Zwiren (4). Briefly, subjects were seated in
their chair with the arm ergometer (Fleisch or Monarch) positioned so that the
axle of the pedal was at shoulder height. Subjects cranked at 50 RPM for 2-min
intervals with increments that would elicit a total test duration of 6-12 min. Test
termination occurred when subjects were unable to maintain the cranking cadence.
Other variables monitored during the test were heart rate (HR) and mechanical
power. HR (bipolar ECG or Sports Tester PE 3000) and RPE were determined
during the last 15 s of each load.
Session 2. At least 2 days later, subjects returned to the lab and were
acquainted with their full range of physical efforts. "Anchoring" of the extremes
of this range was done by repeated 3-min arm cranking bouts at intensities
corresponding to each subject's predetermined lower extreme (RPE 7) and upper
extreme (RPE 19) of the RPE scale. Each of these intensities was performed
three times using a randomized sequence. The subjects rested between bouts
96 - Ward, Bar-Or, Longmuir, and Smith
until their HR recovered to below 100 beats.min-'. Individuals were taught to
"feel" the particular intensity and to recognize it as either the 7 or 19 level.
The same ergometer used for Session 1 was used in Session 2.
Session 3. The following week, subjects reported to the track for the
first prescription session and were instructed to wheel at an even speed for
400 m, using either a 200-m indoor or a 400-m outdoor track (one subject from
each group). Intensities were prescribed as RPE numbers. Subjects performed 4
bouts, one each for RPE 7, 10, 13, and 16. RPE 7 was always prescribed first
to allow for subject warm-up, but was included in the statistical analysis. The
other three tasks were prescribed in a random order. Sufficient rest periods were
given to allow HR to return to <I00 beats.min-I. HR was monitored using the
Sports Tester monitor, with values stored in memory for later recording. A digital
stopwatch was used to record split times at 100-m intervals. Total time was later
used to calculate wheeling speed over the 400-m distance. Subjects used their own
personal wheelchairs, but many of the active subjects had access to specialized
"racing" chairs.
Session 4. Session 4 was intended to assess retention and was conducted
1 month later. Subjects returned to the track and followed a protocol identical
to that of Session 3.
Data Analysis
Differences in subject characteristics between the activity groups were assessed
using a Student's t test. To determine how accurately a subject executed the
prescribed wheeling intensities, individual HR-on-RPE regression lines were
plotted for Session 1 ("criterion session" = C) and for the two prescription
sessions (PI and P2). Group means allowed comparisons of slopes and intercepts
of the three lines by activity status. Data were analyzed using a two-way ANOVA
model, with a Student Neuman-Keuls post hoc test employed whenever F ratio
was significant.
Wheeling velocities and HRs were compared at each of the four prescription
levels (RPE 7,10,13, and 16). Two-way ANOVA with repeated measures (group
and time) was used to analyze retention ability. Tukey's post hoc test was used
to compare group means. An alpha level of .05 was used to determine level of
significance.
Results
Subject characteristics are presented in Table 1. The two groups were similar in
age and weight, but the active group was taller (estimated by arm span) and had
a higher peak aerobic power (per kilogram of body weight) than the sedentary
group (p < .05).
HR on RPE
As a general pattern, HR of all subjects (see Figure I), irrespective of activity
status, increased consistently with increasing RPE number ( p < .0001, F = 23.1,
df = 3). Based on the HR-on-RPE regression lines in sessions C, PI, and P2,
subjects overestimated the choice of wheeling intensities, particularly at the low
Rating of Perceived Exertion - 97
Table 1 Subject Characteristics by Groups
Age
(years)
M
Active (n = 10)
Sedentary (n = 7)
SD
Weight
Ocg)
SD
M
C
skinfolds
(mm)
Arm span
(m>
M
SD
M
SD
Peak
aerobic
power
(Wattlkg)
M
SD
18.8 6.9 55.1 13.3 164.0* 14.9 38.9 17.2 1.33* 0.54
17.7 7.8 51.8 17.2 151.0
6.7 53.4 14.7 0.74 0.41
-
-
0
0
Act,
P1
Act,
P2
Sad. Pl
sad. P2
I
I
I
I
I
Rating of Perceived Exertion
-
Figure 1
Heart rate (HR)at each of the four RPE prescriptions presented as
means f SEM.Each RPE prescription level is significantly higher than the preceding
one 0,< .0001). P1 denotes the first prescription session and P2 the session 1 month
later.
98 - Ward, Bar-Or, Longmuir, and Smith
Rating of Perceived Exertion
Figure 2 - Comparison of all subjects' HR-on-RPE regression lines for the criterion
session and the two prescription sessions (PI and P2). Both P1 and P2 are significantly
different from the criterion session (C) ( p < .05).
efforts. This is displayed in Figure 2, which includes group regression lines for
all subjects combined. The slopes and intercepts for the two prescription tasks
(PI and P2) were significantly different (p < .05) from the criterion (Fintercept
=
4.51, F,,,, = 4.51, df = 2). There was no difference, however, in the degree of
overestimation between the two activity groups. Figure 2 further shows the
similarity of the regression lines in sessions P1 and P2, suggesting a high retention
of a person's ability to execute a prescribed task after one month.
Wheeling Velocities
Velocities chosen by sedentary and active subjects in each of the prescribed RPE
tasks are summarized in Figure 3. At each W E level in either session, active
subjects chose faster velocities than did sedentary subjects (p < .01, F = 14.3,
df = 1). Furthermore, active subjects could better discriminate among the prescribed tasks by progressively selecting higher speeds with increasing RPE numbers ( p < .05, F = 3.57, df = 3). Although active subjects discriminated successfully
between all consecutive RPE prescriptions (7 vs. 10, 10 vs. 13, 13 vs. 16),
sedentary subjects could only discriminate W E 7 from W E 13 and W E 16. No
-
Rating of Perceived Exertion - 99
-
0
Act,
P1
e
Act,
P2
Sad, PI
sed, P2
-
-
-
--
Rating of Perceived Exertion
Figure 3 - Wheeling speed at each of four RPE prescriptions presented as means
f SEM. P1 denotes the first prescription session, P2 the session 1 month later.
Sedentary subjects' speeds are significantly different from those of the trained subjects (p c .01). Wheeling speeds increased significantly (p c .05) for each consecutive
RPE number, except for RPE 10 versus 13 and 13 versus 16 in the sedentary group.
differences were found in either group between the velocities selected 1 month
later. The ability to replicate the performance following a month's interval was
particularly apparent at RPE 7 and 10 for both groups and at RPE 16 for the
active group.
Discussion
The use of a subjective rating scale for the prescription of exercise intensity has
been proposed as an alternative to heart rate counting for use with adults (10,
15, 16). Recent work has established that able-bodied children can discriminate
among various RPE intensity levels in the reproduction of exercise (17, 19, 19).
However, the usefulness of this approach with special populations, such as the
individuals who use wheelchairs, has not been studied.
Data from this study indicate that individuals who use wheelchairs can
discriminate among wheeling intensities as prescribed for them by the Borg
100 - Ward, Bar-Or, Longmuir, and Smith
category scale, but this ability is improved with exercise training. Furthermore,
subjects have displayed an excellent retention of this ability for at least 1 month.
Although subjects could discriminate among RPE prescription levels, HR
for a given RPE level was consistently higher in the prescription phases than in
the criterion sessions. Explanation for this observation is not straightforward.
Arm work uses a smaller muscle mass than does leg or whole body exertion and
could be the main source of the overproduction. One of the few studies that
compared rating of perceived exertion during upper and lower body exercise
found the RPEs were lower for leg than arm exercise (14). However, this study
looked only at estimation of exercise intensity, not production, and observed
responses to prolonged exercise ( 1 0 4 0 min). Another possible explanation for
the overestimation may be related to the use of an arm ergometer for assessing
the criterion and the use of the subject's own wheelchair, freely wheeling, for
the prescription sessions.
Studies with overweight children (17) and healthy adults and children (18)
observed faster speeds and higher HRs during RPE-prescribed outdoor running
when the RPE concept was taught using a cycle ergometer. Ceci and Hassmen
(9) found that physically active adult males tended to execute an RPE prescription
somewhat differently at the outdoor running track than on a treadmill, suggesting
that exercise prescriptions developed from treadmill data be reduced by two RPE
units for use in outdoor settings.
Other studies (5, 12, 16) have found the accuracy of RPE for prescription
to be better at RPE 13 and 16 than at lower intensities, as was noted in this
study. Since this finding was consistent for physical training groups, prescription
of mild to moderate exercise intensities should take into account the higher than
expected HR, and presumably effort, chosen by the individual The following is
an example of the corrections that one could use in prescribing wheeling efforts
to active and less active subjects in order to yield a certain HR:
To achieve HR equivalent to RPE 10, prescribe RPE 6-7.
To achieve HR equivalent to RPE 13, prescribe RPE 9-10.
To achieve HR equivalent to RPE 16, prescribe RPE 14-15.
It was noted (data not included) that younger subjects had greater difficulty at
the lower prescription levels.
Individuals who train in athletic events such as distance running, cycling,
skating, or wheeling are usually able to gauge their velocity. A common practice
in sport training is to describe a certain velocity as a percentage of one's maximal
velocity. It was therefore surprising that the active subjects in this study (those
who exercise on a regular basis) did not show much advantage over the sedentary
group in the accuracy of producing the prescribed tasks. One reason may be the
difference in exercise modality between the practice and the prescription sessions.
Another may be that the particular individuals who took part in this study were
not using fractions of their maximal speed as a means of gauging their wheeling
velocity during training or competition. We are not aware of data on the ability
of healthy athletes to use the RPE as a means of prescription of intensity.
Although training status did not affect a subject's ability to reproduce
exercise intensity, it did alter the pace and range of wheeling speeds. Sedentary
subjects utilized a narrower range of wheeling speeds (0.2-0.3 mls) than did the
Rating of Perceived Exertion - 101
active subjects (0.6-0.7 mfs). It should be noted that many of the active subjects
had access to a different type wheelchair for the prescription sessions. The use
of a specialized ''racing" chair may have contributed to the faster speeds observed
among this group.
The following conclusions can be drawn from this study:
1. Adults and teenagers who rely on a wheelchair as their major means of
transportation can be prescribed wheeling intensities using the Borg category RPE scale.
2. When taught the notion of the scale during an arm-cranking exercise, these
people select wheeling velocities that are higher than expected, based on
their HR-on-RPE regression line as calculated from a maximal ergometer
test. Since overestimation is consistent, the prescription should be adjusted
so that the required HR can be reached.
3. Trained subjects seem to discriminate better among exercise intensities and
utilize a greater range of exercise intensities than do the nonathletes.
4. Such an ability is retained by all subjects, irrespective of activity status,
for at least one month after the original prescription.
5. At any RPE level, trained individuals chose wheeling velocities that are
higher than those chosen by the untrained. However, the HR at each prescribed RPE does not depend on one's training status.
References
1. American College of Sports Medicine. The recommended quantity and quality of
exercise for developing and maintaining cardio-respiratory and muscular fitness in
healthy adults. Med. Sci. Sports Exerc. 22:256-274, 1990.
2. Bar-Or, 0.Pediatric Sports Medicine for the Practitioner. New York: Springer Verlag,
1983.
3. Bar-Or, O., and D.S. Ward. Rating of perceived exertion in children. In: Advances
in Pediatric Sport Sciences: Biological Issues, 0. Bar-Or (Ed.). Champaign, IL:
Human Kinetics, 1989, pp. 151-168.
4. Bar-Or, O., and L. Zwiren. Maximal oxygen consumption test during arm exercise:
Reliability and validity. J. Appl. Physiol. 38:424-426, 1975.
5 . Bayles, C.M., K.F. Metz, R. Robertson, F.L. Goss, J. Cosgrove, and D. McBumey.
Perceptual regulation of prescribed exercise. J. Cardiopulmonary Rehabil. 10:25-31,
1990.
6. Birk, T.J., and C.A. Birk. Use of ratings of perceived exertion for exercise prescription.
Sports Med. 4:l-8, 1987.
7. Borg, G. Perceived exertion as an indicator of somatic stress. Scand. J. Rehabil. 23~92-98,1970.
8. Carton, R.L., and E.C. Rhodes. A critical review of the literature on rating scales for
perceived exertion. Sports Med. 2:198-222, 1985.
9. Ceci, R., and P. Hassmen. Self-monitored exercise at three different RPE intensities
in treadmill vs. field running. Med. Sci. Exerc. Sport 23:732-738, 1991.
10. Chow, R.J., and J.H. Wilmore. The regulation of exercise intensity by ratings of
perceived exertion. J. Cardiac Rehabil. 4:382-387, 1984.
102 - Ward, Bar-Or, Longmuir, and Smith
11. Durnin, J.V.G.A., and J. Womersiey. Body fat assessed from total body density and
its estimation from skinfold thickness: Measurements on 481 men and women from
16-72 years. Br. J. Mutr. 32:77-97, 1974.
12. Eston, R.G., B.L. Davies, and J.G. Williams. Use of perceived effort ratings to control
exercise intensity in young adults. Eur. J. Appl. Physiol. 56:222-224, 1987.
13. Gutmann, M.C., R.W. Squires, M.L. Pollock, C. Foster, and J. Anhotm. Perceived
exertion-heart rate relationship during exercise testing and training in cardiac patients.
J. Cardiac Rehabil. 1:52-59, 1981.
14. Pandolf, K.B., D.S. Billings, L.L. Drolet, N.A. Pimental, and M.N. Sawka. Differentiated ratings of perceived exertion and various physiological responses during prolonged upper and lower body exercise. Eur. J. Appl. Physiol. 535-11, 1984.
15. Pollock, M.J., A.S. Jackson, and C. Foster. The use of the perception scale for exercise
prescription. In: The Perception of Exertion in Physical Work,G. Borg and D. Ottoson
(Eds.). Houndsmills, Basingstoke, UK: Macmillian Press, 1986, pp. 161-176.
16. Smutok, M.A., G.S. Skrinar, and K.B. Pandolf. Exercise intensity: Subjective regulation by perceived exertion. Arch. Phys. Med. Rehabil. 61:569-574, 1980.
17. Ward, D.S., and 0. Bar-Or. Usefulness of Borg Scale in exercise prescription for
overweight youth. Can. J. Sports Sci. 15:120-125, 1990.
18. Ward, D.S., J.D. Jackman, and F. Galiano. Exercise intensity reproduction: Children
versus adults. Ped. Exer. Sci. 3:209-218, 1991.
19. Williams, J.G., R.G. Eston, and C. Stretch. Use of the rating of perceived exertion
to control exercise intensity in children. Ped. Exer. Sci. 3:21-27, 1991.
Acknowledgments
Support for this project was provided by the Ontario Ministry of Tourism and
Recreation. The authors wish to thank Bernd Tragert for his faithful contribution to the
study and Leslie McGillis for her excellent assistance in data collection. Special thanks
go to Marsha Douda, MSPH, for her contribution to the statistical analysis.